Metadata

Filename	GM24385_1.fastq.gz
Filesize	27.16 GiB
Sequali version	1.0.1

Summary

Mean length	9,115.18
Length range (min-max)	3	497,187
Total reads	2,900,987
Q20 reads	0	0.00%
Total bases	26,443,022,791
Total GC bases	10,703,095,094	40.48%
Q20 bases	14,645,135,876	55.38%

Sequence length distribution

See this wikipedia lemma for an explanation about contiguity statistics.

Percentile	Read length
1	18
5	28
10	35
25	54
50	1,039
75	6,442
90	27,121
95	53,491
99	106,963

Contiguity	Read length
N90	6,888
N50	50,349

Per position quality score distribution

Per position quality percentiles (approximation)

Shows the mean for all bases and the means of the lowest and highest percentiles to indicate the spread. Since the graph is based on the binned phreds, rather than exact phreds, it is an approximation.

Per sequence average quality scores

>=Q5	1,403,578	48.38%
>=Q7	1,199,332	41.34%
>=Q10	757,656	26.12%
>=Q12	210,743	7.26%
>=Q15	28	0.00%
>=Q20	0	0.00%
>=Q30	0	0.00%

Per position base content

Per position N content

Per sequence GC content

For short reads with fixed size (i.e. Illumina) the plot will look very spiky due to the GC content calculation: GC bases / all bases. For read lengths of 151, both 75 and 76 GC bases lead to a percentage of 50% (rounded) and 72 and 73 GC bases leads to 48% (rounded). Only 74 GC bases leads to 49%. As a result the even categories will be twice as high, which creates a spike. The smoothened plot is provided to give a clearer picture in this case.

Adapter content

Only adapters that are present more than 0.1% are shown. Given the 12 bp length of the sequences used to estimate the content, values below this threshold are problably false positives. The legend is sorted from most frequent to least frequent.

For nanopore the adapter mix (AMX) and ligation kit have overlapping adapter sequences and are therefore indistinguishable. Please consult the nanopore documentation for more information which adapters are used by your kit.

For illumina short reads, the last part of the graph will be flat as the 12 bp probes cannot be found in the last 11 base pairs.

Adapter counts are accumulated towards the start for front (5') adapters and towards the end for end (3') adapters.

For the reads start and end plots, only adapters found in the first and last 100 bp are shown. All adapter counts are accumulated towards the start and end respectively.

Per tile quality

Per tile quality skipped. Reason: Can not parse header: '35eb0273-89e2-4093-98ed-d81cbdafcac7 runid=1d18d9e9682449156d70520e06571f01c4e6d2d8 sampleid=GM24185_1 read=41492 ch=2628 start_time=2019-01-26T18:52:46Z'.

Duplication percentages

Fingerprints are taken by taking a sample from the beginning and the end at an offset. The samples are combined and hashed while using the length as a seed. A subsample of these fingerprints is stored to estimate the duplication rate. See the documentation for a complete explanation.

Fingerprint front sequence length	8
Fingerprint front sequence offset	64
Fingerprint back sequence length	8
Fingerprint back sequence offset	0
Subsampled fingerprints	533,044
Estimated remaining sequences if deduplicated	72.71%

Overrepresented sequences

A subsample of the sequences is analysed Sequences are cut into fragments of up to 31 bp. Fragments are stored and counted. When the maximum amount of unique fragments is reached, only fragments that are already stored are counted. The rest of the fragments is ignored. Fragments are stored in their canonical representation. That is either the sequence or the reverse complement, whichever has the lowest sort order. Both representations are shown in the table. See the documentation for a complete explanation. The sequence identity is calculated by $\frac{\mathrm{Number\; of\; matched\; nucleotides\; -\; gap\; sequences\; in\; query\; alignment}}{\mathrm{Length\; of\; query}}$.

The percentage shown is an estimate based on the number of occurences of the fragment in relation to the number of sampled sequences. Fragments are only counted once per sequence. Fragments that occur more than once in a sequence are counted as one.

Total sequences in file	2,900,987
Sampled sequences	362,624
Sampling rate	1 in 8
Total fragments sampled	2,600,688
Stored unique fragments	2,281,869
Maximum unique fragments	5,000,000
Fragment size	21

Nanopore time series

Base counts over time

Read counts over time

Active channels over time

Quality distribution over time

Per channel base yield versus quality

translocation speeds

Duration information not available.

Chimeric read splitting

Dorado performs read splitting when it finds adapter sequences in the middle of the signal. The read is split and the original signal uuid is saved in the 'pi' tag.

No 'pi' tags were found.